Working Paper: NBER ID: w30860
Authors: David Hirshleifer; Lin Peng; Qiguang Wang
Abstract: We study how the social transmission of public news influences investors’ beliefs and securities markets. Using data on social networks, we find that earnings announcements from firms in higher-centrality counties generate stronger immediate price, volatility, and trading volume reactions. Post-announcement, such firms experience weaker price drift and faster volatility decay but higher and more persistent volume. These findings indicate that greater social connectedness promotes timely incorporation of news into prices, but also opinion divergence and excessive trading. We propose the social churning hypothesis, which is confirmed using granular data from StockTwits messages and household trading records.
Keywords: social networks; market reactions; earnings announcements
JEL Codes: G11; G12; G14; G4; G41
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Earnings announcements from firms located in higher-centrality regions (R32) | Stronger immediate stock price volatility and trading volume reactions (G14) |
| Earnings announcements from firms located in higher-centrality regions (R32) | Weaker post-announcement drift and faster decay in volatility (G14) |
| Greater social connectedness (Z13) | Timely incorporation of news into prices (G14) |
| Greater social interactions (Z13) | Heavier post-announcement trading but less post-announcement drift (G14) |
| Social connectedness (Z13) | Increased likelihood of household trading (D19) |
| Social connectedness (Z13) | Greater household trading losses (G59) |
| Social interactions (Z13) | Excessive trading and varying levels of market efficiency (G14) |